sign in sign up
<<
Home , 15 Cancri, 20 Cancri, 21 Cancri, 27 Cancri, 28 Cancri, 3 Cancri

SLEUTHING by Guillermo Abramson

ow far away are the ? You might think that shift of the position of a star while the Earth moves along should know, but distances to the stars its . But the stars are so far away that it was only in the Hare something very difficult to figure out. In daily 19th century that astronomers finally succeeded in measur- life, we estimate nearby distances using a trigonometric ing a handful of stellar . Measurements on a grand trick built into our bodies: Our eyes see the world from two scale had to wait for modern technology. slightly different perspectives, and our brain processes this Near the end of the 20th century, the European Space difference to build a three-dimensional image of our envi- Agency (ESA) designed a space to measure stellar ronment. This shift in an object’s apparent position, called parallaxes. The High Precision Collecting Satel- parallax, enables us to complete a myriad of tasks, from lite (Hipparcos, named in honor of the Greek threading a needle to catching a ball in mid-air. Hipparchus of Nicaea from the 2nd century BC), observed a Since classical antiquity astronomers have labored to use predefined set of stars over four . The result was the Hip- the same method on the stars, by observing the apparent parcos Catalogue, published in 1997 and containing precise PLACING the The second data release from the mission solves a decades-long controversy about the distance to the Pleiades cluster.

26 MARCH 2019 • SKY & TELESCOPE parallaxes for a little more than 100,000 stars, all within 300 light-years of Earth. The precision achieved was about 1 milliarcsecond (1 mas), which is 1/1,000 of an arcsecond or 1/3,600,000 of 1°. That is like seeing an astronaut standing on the Moon from Earth. Along with its expected successes, the Hipparcos mis- sion delivered several surprises. The most notable was the distance to a famous stellar cluster, the Pleiades (Messier 45, in ). This group of bluish stars is easily visible by eye in dark autumn skies — although there are many more members than your can discern. Hipparcos found a distance of roughly 380 light-years, rather less than the 440 light-years of previous calculations, which were based on the stars’ brightnesses and considerations of stellar physics (see S&T: June 1999, p. 40). This was an embarrassing problem. On the one hand, if astronomers had been using the wrong distance for the Ple- iades, then it could have implications on a much larger scale: The Pleiades are a nearby and for this reason are frequently used to test our models of p THE PLEIADES Perhaps the most famous in the sky, the bright stars of the Pleiades — often called the Seven Sisters — can be and to calculate the distances to farther stars. On the other seen without binoculars even from a city. It lies roughly 450 light-years hand, if Hipparcos were wrong, then the dubious result might away toward the Taurus. challenge the entire Hipparcos catalog. Was there some NICOLLE R. FULLER / SAYO-ART LLC; PLEIADES PHOTO: DAVIDE DE MARTIN & THE ESA / ESO / NASA PHOTOSHOP FITS LIBERATOR

skyandtelescope.com • MARCH 2019 27 Star Sleuthing

u WHAT IS PARALLAX? Parallax is the shift in an and so on and so forth, changing methods object’s position against the background scene along the way until we reach the confines when viewed from two different locations. Nearby A 2p of the observable universe. Much of mod- stars have measurable parallaxes due to Earth’s motion around the , which astronomers can ern astronomical knowledge, from stellar use to calculate the stars’ distances. However, physics to the structure and evolution of calculating the true distance using the parallax re- the universe, depends on a good calibra- quires eliminating all sources of error in the angle tion of this distance ladder — and thus on measurement. knowing the distances to nearby stars to a T. Controversy over the Pleiades’ distance instrumental or systematic error astrono- was therefore disconcerting. mers had overlooked? Was there a problem p Suspicion quickly fell on Hipparcos, as with just the Pleiades, or also with other additional measurements made with other measurements? Or were the Pleiades really D instruments and methods contradicted closer and thus didn’t fit into our models its result. In 2004, observations of three of stellar formation and evolution? Pleiads done by the Hubble Space Tele- Astronomers now have an answer. scope gave a distance of 435 light-years. In 2014, an extremely precise measurement The Importance of Being made by combining data from radio tele- Clumped scopes all over the world gave a result of Open clusters play a crucial role in astronomy. Because a 444 light-years. Measurement after measurement agreed with cluster’s stars formed together from the same interstellar the greater distance, making the one from Hipparcos look cloud, we know that they are the same age. As such, they anomalously small, even though successive reassessments of are excellent laboratories to test physical models of stellar the spacecraft’s data only narrowed in on the smaller value. evolution. What’s more, by knowing their distance and using It took many years to solve the mystery. The issue is appar- it to derive their intrinsic brightness, astronomers can then ently a matter of instrument calibration, due to Hipparcos’s use these models to calculate the distances to farther stars, intricate observing method. Instead of looking at a fixed spot those that are removed from the reach of direct geometrical in the sky, the telescope rotated about itself and changed the methods like parallax. orientation of its rotation axis over time, a common strategy In this regard, the Pleiades play a keystone role in the cali- for all-sky surveys (see page 20). Using this method, Hip- bration of the cosmic distances ladder, which proceeds step by parcos built up a map of the celestial sphere by determining step, from the Sun to the nearest stars, then to farther stars, the relative angular distances between stars and how those distances changed with time. Calculating a particular star’s parallax required distinguishing the star’s motion from those

T around it. But closely packed stars — like those in a cluster — Gaia DR2 gave tightly correlated measurements. This called for differ- Hipparcos Gaia DR1 VLBI ent calibrations at different spatial scales and resulted in an unexpected source of error for the important and compact open clusters. S But even when they knew what the problem likely was, astronomers had trouble correcting Hipparcos’s data to M produce a distance that agreed with the others. It could be that there were multiple sources of error. So, what scientists wanted was confirmation of the larger distance from an ILLUSTRATION, F instrument that worked as Hipparcos did.

Putting the Pleiades in Their Place B Hipparcos’s successor, the hat-shaped Gaia, provided the opportunity astronomers needed. Like Hipparcos, Gaia is an 2014 ; HIPPARCOS ANOMALY: S&T

ESA satellite of unusual design: It also looks sideways as it S&T

350 400 450 500 GUILLERMO SOURCE: ILLUSTRATION, SCIENCE Distance (light-years) spins, its two scanning the same strip of sky one after the other as the spacecraft slowly rotates its view. Its p HIPPARCOS’S ANOMALY Shown are distances to the Pleiades ac- second data release contains the positions, parallaxes, and cording to a variety of methods. Note the anomalous distance measured by Hipparcos (), the precise result of radio interferometry (green), proper motions of more than 1.3 billion sources and reaches and the early result given by the first Gaia data release (DR1, yellow). The as faint as magnitude 21. The uncertainty of its parallaxes is

article uses Gaia’s second data release (light blue). around 40 microarcseconds (40 μas) for objects brighter than PARALLAX: LEAH TISCIONE / SOURCE: MELIS C. ET AL. / SORTING PARALLAXES: S&T ABRAMSON; GUILLERMO 3D PLOT: ABRAMSON / CELESTIA

28 MARCH 2019 • SKY & TELESCOPE magnitude 15. So while Hipparcos could spot an astronaut on the Moon from Earth, Gaia would be able to see a penny. What value would Gaia find for the distance to the famous cluster? Using the first data release from 2016, one team T M calculated that the 164 cluster stars they included in their A M E analysis gave a distance of 437 light-years. It was a clear con- firmation that Hipparcos’s value was wrong. A How does this analysis work? Using the same method, we can calculate the distance to the Pleiades based on Gaia’s T E second data release, announced in April 2018: First we have to pick the Pleiades out from everything else that shines in the same part of the sky. A download of all the (nearly 700,000!) sources lying within 5° of the Pleiades’ p UNUSUAL STREAM When you plot the positions of the cluster’s members in 3D, the brightest stars form a string stretching away from position gives a cone of observations, with its tip in the solar the swarm and pointing toward the solar system (on the right side of system and extending indefinitely into space. Somewhere this image). The effect might be a byproduct of the data analysis, but inside that cone lie the Pleiades, as well as many field stars in other research has seen similar patterns in open clusters. front of and beyond the cluster. The extraction of the Pleiades from such a large stellar The Pleiades have not relinquished all their mysteries, population is, fortunately, quite simple. The stars of an open however. The precise determination of the positions allows cluster move together and therefore share the same proper us to plot the cluster in 3D. When doing so we can see that motion across the sky. Gaia records a for the the brightest stars (the traditional Seven Sisters, with proper Pleiades of around 6 mas per , in a particular southeast- names after the Pleiades of mythology) form a string stretch- ern direction. This leaves 1,876 stars of our original 700,000. ing away from the swarm and pointing toward our solar sys- Almost all of them belong to the cluster. tem. This is, frankly, bizarre. What are the chances that the If we sort these remaining stars based on their parallaxes, brightest, most massive stars are not only on the same side of we find that the vast majority are clumped together, with an the cluster but form a tail? average shift in their apparent positions of 7.3 mas. Selecting It’s highly likely that this feature is an artifact of the those stars with parallaxes between 5 and 9.5 leaves 1,594 measurement. The parallaxes of these brightest stars have stars, most of them surely belonging to the cluster. When particularly large errors, because Gaia is fine-tuned for stars we factor in various sources of error, we are left with a very between magnitudes 6 and 15, and the Seven Sisters are precise final result of .7 317 ± 0.002 mas, which corresponds much brighter. It’s also possible that at least some of these to a distance of 445.8 ± 0.1 light-years. bright stars are not actually part of the cluster: The closest It comes almost as a relief: The Pleiades are where they ones, and , are moving faster across the sky than should be, and stellar physics is all right. the bulk of the cluster’s members. Then again, it could be that the cluster is simply elongated. And it’s not a wild idea: Other research has seen similar 700 patterns in open clusters across the Milky Way. We can only hope that further analysis and refinements in the catalog will 600 shed light on these matters. 500 ¢ GUILLERMO ABRAMSON is a physicist at the Statistical and 400 Interdisciplinary Physics Group of the Bariloche Atomic Center ILLUSTRATION, and Balseiro Institute, in Bariloche, Argentina. This work made 300 use of Gaia mission data processed by the Gaia Data Process-

CELESTIA ing and Analysis Consortium.

Number of objects 200

100 FURTHER READING: To learn the fascinating history of stel- 2014 F lar parallax, read Parallax: The Race to Measure the Cosmos, by ; HIPPARCOS ANOMALY: S&T 0 Alan W. Hirshfeld (W. H. Freeman, 2001). “Hipparcos: The Stars ILLUSTRATION, SOURCE: GUILLERMO GUILLERMO SOURCE: ILLUSTRATION,

S&T 0 1 2 3 4 5 6 7 8 9 10 in Three Dimensions” (S&T: June 1999, p. 40) recounts the results Parallax (milliarcseconds) of the Hipparcos mission, including the mystery of the Pleiades. The paper “Gaia Data Release 2: Using Gaia Parallaxes” by Xavier p SORTING PARALLAXES There are nearly 700,000 sources that lie within 5° of the Pleiades’ position. To tease out the stars that belong to Luri (University of Barcelona, Spain) and colleagues (Astronomy & the cluster, we plot all the sources’ parallaxes. The cluster’s members Astrophysics 2018) discusses the methods to obtain distance from PARALLAX: LEAH TISCIONE / S&T SOURCE: MELIS C. ET AL. / SCIENCE ABRAMSON; GUILLERMO 3D PLOT: ABRAMSON / SORTING PARALLAXES: clump together. parallax in the context of Gaia DR2.

skyandtelescope.com • MARCH 2019 29